The past decades have witnessed an explosive growth of the Internet Things (IoT) services requiring intensive computation resources. conventional IoT devices, however, are usually equipped with very limited resources, which results in degraded quality experience when executing resource-hungry applications. Mobile edge computing (MEC), enables smart terminals (STs) to offload parts their workloads servers located at cellular base stations (BSs), has provided a promising approach address this...

Driven by the explosive growth in computation-intensive applications future 5G networks and industries, mobile edge computing (MEC), which enables smart terminals (STs) to offload their computation workloads nearby servers (ESs) radio access networks, has attracted increasing attention. In this article, we investigate energy-efficient multitask multiaccess MEC via nonorthogonal multiple (NOMA). Exploiting NOMA, an ST with tasks can respective of different ESs simultaneously. To study...

In this paper, a novel hybrid Neural Network Algorithm-Differential Evolution (NNA-DE) optimizer which integrates both NNA and DE is proposed. The proposed NNA-DE demonstrates better performance compared to the standard other state-of-the-art optimization algorithms. algorithm then employed for optimizing an observer-based interval type-2 fuzzy PID (OB-IT2FPID) controller applied proton exchange membrane fuel cell (PEMFC) air feeding system. whole design parameters of OB-IT2FPID including...

The quality of unmanned surface vehicle (USV) local path planning directly affects its safety and autonomy performance. USV might easily be trapped into optima. swarm intelligence optimization algorithm is a novel effective method to solve the path-planning problem. Aiming address this problem, hybrid bacterial foraging with simulated annealing mechanism proposed. proposed preserves three-layer nested structure, incorporated outermost dispersal operator. can effectively escape Convention on...

Abstract High-dimensional optimization has numerous potential applications in both academia and industry. It is a major challenge for algorithms to generate very accurate solutions high-dimensional search spaces. However, traditional tools are prone dimensional catastrophes local optima, thus failing provide high-precision results. To solve these problems, novel hermit crab algorithm (the HCOA) introduced this paper. Inspired by the group behaviour of crabs, HCOA combines optimal historical...

The blood oxygen level-dependent (BOLD) signal derived from functional neuroimaging is commonly used in brain network analysis and dementia diagnosis. Missing the BOLD may lead to bad performance misinterpretation of findings when analyzing neurological disease. Few studies have focused on restoration time-series data.

Carbon emissions play a significant role in shaping social policy-making, industrial planning, and other critical areas. Recurrent neural networks (RNNs) serve as the major choice for carbon emission prediction. However, year-frequency data always results overfitting during RNN training. To address this issue, we propose novel model that combines oscillatory particle swarm optimization (OPSO) with long short-term memory (LSTM). OPSO is employed to fine-tune hyperparameters of LSTM, utilizing...

Abstract Assistive medical image classifiers can greatly reduce the workload of personnel. However, traditional machine learning methods require large amounts well-labeled data and long times to solve classification problems, which lead high training costs poor applicability. To address this problem, a novel unsupervised breast cancer model based on multiscale texture analysis dynamic strategy for mammograms is proposed in paper. First, gray-level cooccurrence matrix Tamura coarseness are...

A scheme to solve the course keeping problem of unmanned surface vehicle with nonlinear and uncertain characteristics unknown external disturbances is investigated in this article. The chattering existing global fast terminal sliding mode controller solving disturbance analyzed. To reduce eliminate influence disturbance, an adaptive based on radial basis function neural network developed. equivalent control that usually requires a precise model information system computed using network....

A twinning bare bones particle swarm optimization(TBBPSO) algorithm is proposed in this paper. The TBBPSO combined by two operators, the twins grouping operator (TGO) and merger (MO). TGO aims at reorganization of swarm. Two particles will form as a twin influence each other subsequent iterations. In twin, one designed to do global search while local search. MO merging enhancing ability main group. operators work together enhance minimum escaping methods. addition, no parameter adjustment...

The obstacles modeling is a fundamental and significant issue for path planning automatic navigation of Unmanned Surface Vehicle (USV). In this study, we propose novel method based on high resolution satellite images. It involves two main steps: extraction obstacle features construction convex hulls. To extract the features, series operations such as sea-land segmentation, details enhancement, morphological transformations are applied. Furthermore, an efficient algorithm proposed to mask...

The unmanned surface vessel (USV) trajectory with spatial and temporal information plays an important role in its positioning navigation. Unlike traditional reconstruction methods, this paper proposes a novel method based on the automatic identification system (AIS) for USV. Aside from AIS data applied restoring USV's trajectory, proposed considers constraints of vessel's navigation state, maneuvering factors time stamps. This consists three steps: restoration, Empirical Mode Decomposition...

The offspring selection strategy is the core of evolutionary algorithms, which directly affects method's accuracy. Normally, to improve search accuracy in local areas, population converges quickly around optimal individual. However, excessive aggregation can narrow range population, and thus may be trapped by optima. To overcome this problem, a bare-bones particle swarm optimization with crossed memory (BPSO-CM) proposed work. BPSO-CM contains multi-memory storage mechanism (MSM) an elite...

Unmanned surface vehicle has the properties such as complexity, nonlinearity, time variability, and uncertainty, which lead to difficulty of obtaining a precise kinematics model. A neural adaptive sliding mode controller for unmanned steering system is developed based on control technique radial basis function network. In new approach, two parallel networks are used reduce influence uncertainties eliminate dependency model system. Among these networks, one approximate unknown nonlinear yaw...

When an unmanned surface vehicle (USV) navigates in narrow waterway scenarios, its ability to detect vanishing points accurately and quickly is highly important for safeguarding navigation safety realizing automated navigation. We propose a novel approach detecting based on improved lightweight AlexNet. First, similarity evaluation calculation method image texture features proposed, by which some scenarios are selected from the filtered Google Street Road Dataset (GSRD). These together with...

Significant wave height (SWH) prediction is crucial for marine safety and navigation. A slow failure particle swarm optimization long short-term memory (SFPSO-LSTM) proposed to enhance SWH accuracy. This study utilizes data from four locations within the EAR5 dataset, covering 1 January 31 May 2023, including variables like wind components, dewpoint temperature, sea level pressure, surface temperature. These predict at 1-h, 3-h, 6-h, 12-h intervals. SFPSO optimizes LSTM training process....

Abstract Recognition of obstacle type based on visual sensors is important for navigation by unmanned surface vehicles (USV), including path planning, avoidance, and reactive control. Conventional detection techniques may fail to distinguish obstacles that are similar in appearance a cluttered environment. This work proposes novel recognition approach combines dilated operator with the deep-level features map ResNet50 autonomous navigation. First, images collected annotated from various...

Abstract To solve the long-tail problem and improve testing efficiency for autonomous navigation systems of unmanned surface vehicles (USVs), a visual image-based scene complexity perception method is proposed. In this paper, we intend to accurately construct mathematical model between features from analysis processing image textures. First, typical complex elements are summarized, scenes divided into four levels according whether they contain these elements. Second, textural extracted using...